支持向量机最优模型选择的研究

Optimal Model Selection for Support Vector Machines

通过对核矩阵的研究,利用核矩阵的对称正定性,采用核校准的方法提出了一种SVM最优模型选择的算法———OMSA算法.利用训练样本不通过SVM标准训练和测试过程而寻求最优的核参数和相应的最优学习模型,弥补了传统SVM在模型选择上经验性强和计算量大的不足.采用该算法在UCI标准数据集和FERET标准人脸库上进行了实验,结果表明,通过该算法找到的核参数以及相应的核矩阵是最优的,得到的SVM分类器的错误率最小.该算法为SVM最优模型选择提供了一种可行的方法,同时对其他基于核的学习方法也具有一定的参考价值.

Proposed in this paper is a method of model selection based on kernel alignment for support vector machines-OMSA (optimal model selection algorithm) by means of learning on kernel matrix. This algorithm aims at finding the optimal kernel parameters and learning model from training data without performing the standard procedures of SVM training and testing so as to overcome the flaws of conventional methods of SVM model selection. The classification experiments on the UCI database and the face recognition experiments on the FERET face database are deployed with this algorithm and the famous LOO (leave-one-out) algorithm. The four datasets from UCI used in experiments are diabetis, glass, waveform and wine. By comparison with the LOO algorithm, the experimental results show that the optimal kernel parameters and kernel matrix are found by OMSA with the minimal testing error of SVM classifier. Specially, the results from face recognition experiments are satisfactory. This algorithm provides a feasible method for SVM model selection as well as references for other kernel-based learning algorithms.